The Lyme Bomb Detector is the first smartphone tool enabling earlier detection of Lyme rash on Black and Brown skin and more accurate diagnosis of the disease overall.
This tool is fueled by artificial intelligence that allows users to check their skin for rashes and assess their risk. A single picture is analyzed using a neural network model which has been trained with photographs showcasing different skin tones and different variations of visual symptoms.
Currently, the Lyme Bomb Detector prototype is achieving 79.6% accuracy in detecting Lyme rashes from smartphone pictures. But more photos are being added to continue training the tool.
Lyme disease is a chronic condition caused by bacteria that are transmitted by ticks. Long-term effects of the disease can be widespread and debilitating.
Ticks are quickly moving from the farms and woods to our neighborhoods, and Black and Brown urban communities are most at risk because the tell-tale sign of a tick bite—the red rash—is harder to spot on dark skin. So that population often ends up misdiagnosed and with untreated, severe Lyme disease.
Making matters worse, almost no documentation exists currently in textbooks or online to help doctors and patients recognize rashes on darker skin tones. In an effort to raise awareness among city-dwellers and improve Lyme diagnosis rates in people of color, the Lyme Bomb Detector was created.